'pigment epithelium of eye' logical definition does not match textual definition

[balhoff]

The definition of pigment epithelium of eye is "Epithelial layer of the retina, ciliary body, or iris composed of cells containing pigment granules."

However the logical definition is simply epithelium and ('part of' some eye).

As a result eye epithelium is inferred to be a subclass of pigment epithelium of eye, which doesn't sound right to me.

[dosumis]

@dragon-ai-agent please relax the logical definition of this term to subclassof axioms.

[dosumis]

@balhoff - attempting an AI edit. But I'm not really sure why this term exists. Maybe better to obsolete it?

[cmungall]

I've relaxed the logical definition of 'pigment epithelium of eye' (UBERON:0007625) by replacing the intersection_of axioms with direct is_a and part_of relationships. This changes the logical definition from:

epithelium and ('part of' some eye)

to simple subclass axioms:

is_a: epithelium
relationship: part_of eye

This should prevent the incorrect inference that all eye epithelium is a pigment epithelium. Pull request created: #3538

Signed-off-by: @dragon-ai-agent

[balhoff]

@balhoff - attempting an AI edit. But I'm not really sure why this term exists. Maybe better to obsolete it?

I don't feel informed enough to say. At least relaxing the logical definition is a good start.

[github-actions]

This issue has not seen any activity in the past 6 months; it will be closed automatically one year from now if no action is taken.